LIANA ABUNDANCE AND FUNCTIONAL DIVERSITY ALONG AN ALTITUDINAL GRADIENT IN NORTHERN ECUADOR Word count: 30 564 Camille Meeussen Student number: 01200362 Supervisor: Prof. dr. ir. Hans Verbeeck Co-supervisor: dr. ir. Elizabeth Kearsley A dissertation submitted to Ghent University in partial fulfilment of the requirements for the degree of Master in Bioscience Engineering: Forest and Nature Management Academic year: 2016 - 2017 ii The author and supervisors give the permission to use this thesis for consultation and to copy parts of it for personal use. Every other use is subject to the copyright laws, more specifically the source must be extensively specified when using results from this thesis. De auteur en promotors geven de toelating deze scriptie voor consultatie beschikbaar te stellen en delen ervan te kopiëren voor persoonlijk gebruik. Elk ander gebruik valt onder de beperkingen van het auteursrecht, in het bijzonder met betrekking tot de verplichting uitdrukkelijk de bron te vermelden bij het aanhalen van resultaten uit deze scriptie. Ghent, June 2017 Prof. dr. ir. Hans Verbeeck dr. ir. Elizabeth Kearsley Camille Meeussen iii Acknowledgements I have never been good in making decisions. I always considered it as missing out on something, missing out on an experience, a chance to meet new people or an opportunity to learn. But the final years I only made the best possible decisions in my life. I went to study bio-engineering at the most beautiful faculty of Ghent, surrounded by friendly and sincere people. I couldn’t be more right choosing Forest and Nature Management as my master degree and finally I also selected a wonderful thesis subject. I’m very grateful for the people I’ve met, the friends I’ve made, the places I’ve visited and the things I’ve learned (although I’m definitely not satisfied yet). So, some acknowledgements are certainly in place. More specifically I really want to express my gratitude to the persons supporting and supervising me during these final years. First of all, I want to thank my supervisor professor Hans Verbeeck. I really appreciate you gave me this opportunity but I also want to thank you for the time you made for me throughout the year and for reviewing my thesis. Additionally, I want to express my great gratitude to my co-supervisor Elizabeth Kearsley. Thank you for your comments and corrections, for your support and at the same time the freedom I was given during the whole process of investigation and writing this thesis. Elizabeth, thank you very much for reviewing and improving my work. I also want to thank Marijn Bauters for making time to explain and guide me through the chemical analyses. Furthermore, I want to thank Katja Van Nieuland and Stijn Vandevoorde for completing the chemical analyses and providing me with all the data I needed to finalize this thesis. Debbie, thank you for taking care of us at the other side of the world and for the preparations and the arrangements. You were a big support for us, especially during our first days in Ecuador! I want to thank Niko for the time he dedicated to this liana inventory. He didn’t only made time for the field campaign and for the determination of the different liana species. But also when I was back in Belgium he was always there to answer my questions. Niko, I also want to thank you for inspiring me with your love and knowledge of Ecuador’s wonderful nature, for your everlasting enthusiasm and making me feel like home. Thank you for adopting my invented Spanish words and only telling me after eight weeks they didn’t even exist. Hopefully one day we will meet again. I would like to thank Fausto as well for the helping hand during the inventories, the same applies for Andres. Furthermore, I would like to thank Evelien. Thank you for going with me on this adventure and for sharing with me the ups and downs. I think we will never forget that Ecuador is the country with the most dangerous cows, the country where for some people skin colour matters more than having brains and the country with an awful lot of eggs and rice. But at the same time, we will always remember Ecuador as the county with the most magnificent nature and the most marvellous animals, the best busses and the breathtaking views. I really had a lovely experience working and travelling with you at the other side of the world! When do we return? Finally, I want to thank my friends and family for their love and support and for the happiness they bring into my life. A special thanks to my mom, for the endless support and trust you have in me! Camille Meeussen, June 2017 iv Table of contents 1. INTRODUCTION .................................................................................................................... 1 2. LITERATURE REVIEW ............................................................................................................ 3 2.1. Global distribution of vegetation ......................................................................................................... 3 2.1.1. The tropics ......................................................................................................................................... 3 2.1.2. Tropical montane cloud forest .......................................................................................................... 4 2.2. Global trends in density and diversity of lianas.................................................................................... 5 2.2.1. Factors influencing global liana distribution ..................................................................................... 6 2.2.1.1. Temperature ................................................................................................................................. 6 2.2.1.2. Soil nutrients ................................................................................................................................. 7 2.2.1.3. Mean annual precipitation and seasonality .................................................................................. 7 2.2.2. Factors influencing global liana species richness .............................................................................. 7 2.3. Altitudinal gradient.............................................................................................................................. 8 2.3.1. Vegetation along an altitudinal transect in the tropics ..................................................................... 9 2.3.2. Lianas along an elevational transect ............................................................................................... 10 2.3.3. Factors controlling lianas along elevational gradients .................................................................... 10 2.4. Changing environment ...................................................................................................................... 11 2.4.1. Climate change ................................................................................................................................ 11 2.4.2. Changing tropical forests................................................................................................................. 12 2.4.3. Liana proliferation ........................................................................................................................... 13 2.4.3.1. Factors causing liana proliferation .............................................................................................. 13 2.4.3.2. Effects of liana proliferation ....................................................................................................... 15 2.5. Biological diversity ............................................................................................................................. 16 2.5.1. Liana functional traits ...................................................................................................................... 16 2.5.1.1. Seeds, flowers and fruits ............................................................................................................. 16 2.5.1.2. Leaf traits .................................................................................................................................... 17 2.5.1.3. Climbing mechanism ................................................................................................................... 18 2.5.1.4. Wood traits ................................................................................................................................. 18 2.5.1.5. Root traits ................................................................................................................................... 19 2.5.1.6. Biogeographical variation in liana traits ..................................................................................... 19 2.5.1.7. Latitudinal variation in liana traits .............................................................................................. 19 3. MATERIAL AND METHODS ................................................................................................. 21 3.1. Study area and plot description ......................................................................................................... 21 3.2. Data acquisition ................................................................................................................................. 24 3.3. Leaf sampling and analysis ................................................................................................................ 25 3.4. Statistical analysis.............................................................................................................................. 26 4. RESULTS ............................................................................................................................. 29 v 4.1. Liana community structure ................................................................................................................ 29 4.2. Liana species diversity and functional diversity ................................................................................. 31 4.2.1. Taxonomic diversity ........................................................................................................................ 31 4.2.2. Liana leaf traits ................................................................................................................................ 33 4.2.3. Functional diversity ......................................................................................................................... 37 4.3. Functional community structure ........................................................................................................ 38 5. DISCUSSION ....................................................................................................................... 43 5.1. Liana community structure along the altitudinal gradient ................................................................. 43 5.2. Species diversity along the altitudinal gradient ................................................................................. 45 5.3. Liana functional diversity along the altitudinal gradient .................................................................... 46 5.3.1. Liana leaf traits ................................................................................................................................ 46 5.3.2. Functional diversity ......................................................................................................................... 47 5.4. Functional community structure of the forest along the altitudinal gradient ..................................... 48 6. CONCLUSION ..................................................................................................................... 51 7. REFERENCES ....................................................................................................................... 53 8. APPENDIX ........................................................................................................................... 69 vi Abbreviations a.s.l.: Above sea level AGB: Aboveground biomass BA: Basal area CWM: Community weighted mean DBH: Diameter at breast height ENSO: El Niño Southern Oscillation FAO: Food and Agricultural Organisation of the United States FDis: Functional dispersion FDiv: Functional divergence FEve: Functional evenness FRic: Functional Richness LCC: Mass-based leaf carbon content LES: Leaf economic spectrum LNC: Mass-based leaf nitrogen content LPC: Mass-based leaf phosphorus content MAP: Mean annual precipitation MAT: Mean annual temperature MCF: Mindo Cloud forest Foundation MST: Minimum spanning tree Ntot: Total nitrogen content Pbio-av: Bioavailable phosphorus PC: Principal component PCA: Principal component analysis PSP: Permanent sample plot Ptot: Total phosphorus content SLA: Specific leaf area TMCF: Tropical montane cloud forest VPD: Vapor pressure deficit WUE: Water use efficiency vii Summary During the last decades, an increase in liana abundance and biomass in Neotropical forests has been noted (Phillips et al., 2002a; Schnitzer and Bongers, 2011). Lianas are an important component of tropical forests, contributing a lot to its structure and biodiversity (Gentry, 1991; Chave et al., 2001), but, more specifically liana competition also impacts tree mortality, growth, and diversity, among others (Schnitzer and Bongers, 2002; van der Heijden et al., 2015). Lianas therefore affect the carbon balance and dynamics and thus changes in the liana community are of particular interest (van der Heijden et al., 2013, 2015). The main target of this study was to investigate the shift in liana community structure and functional identity along an altitudinal gradient. Such an elevational gradient is an ideal setup, as on a small spatial scale, changes in abiotic and biotic factors can be linked with the changing vegetation. In particular, in this thesis we investigated lianas (diameter ≥ 2 cm) along an altitudinal gradient (400 m a.s.l.-3,200 m a.s.l.) established on the western slope of the Andes in northern Ecuador. All 17 permanent sample plots (40 x 40 m) are located in natural undisturbed old-growth tropical lowland or montane forest divided over four different altitudes. The relationship between the liana community structure (liana diameter distribution, density, AGB and BA) and the altitude was explored, to find out if we could observe a shift in liana community structure with elevation. Additionally, also changes in liana species diversity, functional diversity and leaf traits (SLA, LNC, LPC, LCC, δ13C, δ15N, C:P, C:N and N:P) along the altitudinal gradient were studied. Finally, a comparison was made between the leaf traits of lianas and co-occurring trees to examine if both growth forms utilize the same functional niche or show functional dissimilarities. Average liana diameter decreases significantly with increasing altitude probably driven by a reduction in temperature. Liana density peaks at 1,900 m a.s.l. but does not show a trend across the different strata. Hence, we can detect a marginal significant decrease in liana AGB and BA. In general, we cannot report a very strong influence of soil fertility on liana density or AGB, only the soil C:N ratio shows a significant negative correlation with liana diameter, liana AGB and liana BA. Not all liana species were identified but almost all species are site-specific and only two species occur in more than one stratum. Further every altitude is dominated by different families. This turnover of species along the altitudinal gradient was also reflected in a change in functional leaf traits. There is a clear shift from the quick return end of the leaf economic spectrum towards a slower return on investment up higher in the mountain chain, with a significant decrease in the CWM of SLA, LPC, δ15N and LNC and a significant increase in LCC. This changing strategy is a consequence of a shift towards harsher conditions. Namely, a decrease in temperature, precipitation and soil nutrients availability, forces the lianas to switch towards a more conservative strategy. The same trends are also noticed in the leaves of the trees in the same plots but these trees tend to have lower values for SLA, LNC and LPC compared to the lianas. This indicates a dissimilarity in functional niche between the two growth forms whereby lianas are able to maintain a better resource acquisition and growth rate in comparison with the co-occurring trees. Furthermore, no reduction in niche space for lianas is found across the gradient, as deduced from functional diversity indices, while this was the case for the co-occurring trees. viii Samenvatting De voorbije decennia is er een toename in densiteit en biomassa van lianen waargenomen in tropische bossen in de Neotropen (Phillips et al., 2002a; Schnitzer en Bongers, 2011). Lianen zijn belangrijke componenten van deze bossen, sterk bijdragend aan hun structuur en biodiversiteit (Gentry, 1991; Chave et al., 2001), maar toch moet deze trend nauwlettend opgevolgd worden omwille van de competitie tussen lianen en bomen. Er is een impact op onder andere boom mortaliteit, groei en diversiteit (Schnitzer en Bongers, 2002; van der Heijden et al., 2015). Zo zal ook de koolstofbalans en - dynamiek worden beïnvloed (van der Heijden et al., 2013, 2015). Het voornaamste doel van deze thesis is het onderzoeken van veranderingen in het voorkomen en in de functionele identiteit van lianen langsheen een hoogtegradiënt. Hoogtegradiënten zijn de ideale proefopzet om op een kleine afstand het effect van veranderende abiotische en biotische factoren te kunnen linken met wijzigingen in de vegetatie. In deze thesis worden lianen (diameter ≥ 2 cm) onderzocht langsheen een hoogtegradiënt (400-3200 m boven zeeniveau) op de westerflank van het Andesgebergte in het noorden van Ecuador. Alle 17 permanente proefvlakken (40 x 40 m) zijn gelegen in natuurlijk, onverstoord bos, behorend tot het tropisch laaglandregenwoud of bergnevelwoud, verdeeld over vier hoogtes. De relatie tussen de diameter, densiteit, AGB en BA van de lianen en de hoogte wordt onderzocht om eventuele trends te ontdekken. Vervolgens wordt ook gekeken naar veranderingen in soortenrijkdom, functionele diversiteit en bladkenmerken (SLA, LNC, LPC, LCC, δ13C, δ15N, C:P, C:N en N:P) van lianen langsheen de gradiënt. Uiteindelijk wordt ook een vergelijking gemaakt tussen de bladkenmerken van lianen en bomen om na te gaan of deze twee groeivormen dezelfde functionele niche bezetten of eerder functioneel verschillend zijn. Langsheen de gradiënt daalt de gemiddelde diameter van de lianen significant met de hoogte, waarschijnlijk een gevolg van de dalende temperatuur. De densiteit is het grootst op een hoogte van 1900 meter maar vertoont geen significante verschillen met de andere strata. Hieruit volgend kunnen we een zwak dalende trend in AGB en BA waarnemen met stijgende hoogte. Over het algemeen kunnen we geen sterke invloed van bodemnutriënten op de densiteit en AGB van de lianen opmerken, enkel de verhouding C:N van de bodem is gecorreleerd met de diameter, AGB en BA van lianen. Niet alle individuen werden geïdentificeerd maar de meeste soorten komen enkel voor in 1 stratum, slechts 2 soorten vormen hierop een uitzondering. Verder wordt iedere hoogte ook gedomineerd door andere families. Deze turnover wordt ook gereflecteerd in een verandering in bladkenmerken. Er is een duidelijk daling waarneembaar van het quick return end van het LES naar een slower return on investment hogerop in de bergen, met een significante daling in CWM van de SLA, LPC, LNC en δ15N en een significante stijging in LCC. Deze verandering in strategie is het gevolg van een overgang naar extremere omstandigheden. Namelijk een daling in temperatuur, neerslag en nutriënten beschikbaarheid dwingen de lianen over te schakelen naar een meer conservatieve strategie. Dezelfde trend is ook opgemerkt in de bladkenmerken van bomen in dezelfde proefvlakken, hoewel de gemiddelde waarden voor de SLA, LNC en LPC wel lager zijn voor de bomen in vergelijking met de lianen. Dit wijst op het gebruik van een verschillende functionele niche voor beide groeivormen waarbij de lianen over het algemeen efficiënter nutriënten kunnen opnemen en een betere groei kunnen behouden. Verder is ook geen reductie waargenomen in functionele niche voor de lianen langsheen de gradiënt terwijl dit wel zo is voor de bomen. ix 1. Introduction Woody vines or lianas are plants that root in the soil and use other plants, mostly trees, to grow towards the canopy. They are structural parasites making use of a host tree to reach better light levels (Schnitzer and Bongers, 2002). These lianas are an important life form in tropical forests. They profoundly contribute to biodiversity by representing 25-35% of the woody species diversity in tropical forests (Chave et al., 2001; Schnitzer et al., 2012). Furthermore, lianas play a significant role in the carbon sequestration and stock by encompassing approximately 10-40% of the woody stems (Gentry, 1991; Chave et al., 2001) and even up to 40% of the leaves present in the forest can be assigned to the lianas (Putz, 1984a). Although lianas are an important component of the ecosystem, they can severely impact carbon forest dynamics too. Once established in the upper canopy they shade the leaves of their hosts (Zhu and Cao, 2009), thereby creating severe aboveground competition with trees for light. Additionally, lianas have particularly deep rooting systems (Restom and Nepstad, 2004), invoking strong belowground competition for nutrients and water. This strong competition from lianas reduces the growth of trees and is responsible for a higher risk of tree mortality (Putz, 1984a; reviewed by Schnitzer and Bongers, 2002; Ingwell et al., 2010). Furthermore, tree recruitment, fecundity and tree diversity can also be affected (van der Heijden et al., 2015). In this way lianas are able to influence the carbon sequestration in tropical forests (van der Heijden et al., 2013, 2015). According to Pan et al. (2011) 55% of the carbon stored in forests is attributed to the tropics and thus changes in the dynamics and productivity of these forests may have consequences for global CO levels (van der Heijden et al., 2013). 2 The impact of lianas on the forest carbon balance can potentially increase, with a rise in liana density and biomass recorded in Central and South America (Phillips et al., 2002a; Schnitzer and Bongers, 2011). Possible drivers of this increase are human disturbance, higher evapotranspirative demand and a rise in atmospheric CO (Schnitzer and Bongers, 2002, 2011). Lianas seem to profit from these 2 situations because of their efficient water use and multiple ways of reproduction (Schnitzer, 2005). Furthermore, they invest less biomass in their stem (Schnitzer et al., 2000) and can produce more leaves with a low investment cost (van der Sande et al., 2013). Lianas are therefore causing more pressure on the forests (Schnitzer and Bongers, 2002) and they will possibly acquire the ability to spread towards higher altitudes and latitudes (Jiménez-Castillo et al., 2007). Within this dissertation, a first liana census was carried out in 17 permanent sample plots in northern Ecuador. These plots are situated in four strata along an altitudinal gradient (400-3,200 m above sea level (a.s.l.)) on the western slope of the Andes. Elevational transects are an invaluable source of information in the investigation of climate change. On a small spatial scale, changing abiotic factors can be studied and they can be linked with the surrounding vegetation (Körner, 2007). Especially the gradual shift in temperature serves as an indicator of climate change. The purpose of this study is twofold. The first part is to investigate if there is a shift in liana community structure variables such as liana density, diameter distribution, basal area and aboveground biomass along the elevational transect. Based on recent literature on lianas along altitudinal gradients (Parthasarathy et al., 2004; Fadrique and Homeier, 2016) we expect to see a decrease in liana diameter, density and biomass as temperature decreases higher in the Andes. Namely, lianas are sensitive to freezing induced embolism which limits also their latitudinal dispersion (Schnitzer and Bongers, 2002; Jiménez-Castillo et al., 2007). The second part of the study focusses on changes in species richness and functional diversity along the same gradient. Also here we assume to find a drop in diversity with increasing altitude because of a shift towards harsher growth conditions (van der Heijden and Phillips, 2009a; DeWalt et al., 2015). Leaf traits of lianas, namely the specific leaf area (SLA), mass-based leaf carbon content 1