National Science Foundation University of Connecticut PBDFF Louisianna State Universtiy CIECO Centro Agronomico Tropical de Investigacion y EzsenanzaWageningen Organization for Tropical Studies

ATBC2012

Abstracts Symposium of the Association of Tropical Biology and Conservation meeting in June 2012, Bonito, Brazil

 

(1) Plant functional traits and the slow-fast continuum along successional gradients of Mexico: Do dry and wet tropical forests show parallel trait continua?

Lohbeck M2,1, Tauro A1, van den Elzen E, Meave J3, Lebrija-Trejos E4, Romero E3, Poorter L2, Martínez-Ramos M1, Paz H1, Bongers F21Centro de Investigaciones en Ecosistemas, UNAM, Mexico, 2Wageningen University, the Netherlands – Forest Ecology and Forest Management group, 3UNAM, Mexico – Departamento de Ecología y Recursos Naturales, 4Smithsonian Tropical Research Institute

 

Succession is community assembly in action, and functional traits are thought to drive changing community composition during succession. We test whether there is a change from ‘fast’ acquisitive traits early in succession towards ‘slow’ conservative traits later in succession, and whether forest structure (stand basal area as an indicator of speed of succession and of environmental conditions) is a better predictor of these patterns than time since abandonment. Research was carried out in 17 wet secondary forest sites (0- 25yrs) and 16 dry secondary forest sites (5-63 yrs) in Mexico. Functional characteristics of the plant community were described with the ‘community-weighted mean’(CWM; average trait value of the overall community), which was calculated based on 12 functional traits and a total of 130 species. Stand basal area and time since abandonment predicted changes in CWM equally well in both forest types. Traits are more strongly related to succession in dry forest (explaining 8 traits) than in wet forest (explaining 6 traits). Specific leaf area, a ‘fast’ trait reflecting the efficiency of leaf display decreased, and biotic seed dispersal increased with succession in both dry and wet forest. Leaf tissue density and leaf dry matter content increased during wet forest succession, in line with the hypothesis. Surprisingly, the ‘slow’ traits leaf tissue density, leaf dry matter content, and wood density decreased during dry forest succession, in contrast to the hypothesis. Probably these conservative ‘slow’ traits are important in early dry forest succession because they are related to drought resistance. We conclude that secondary forests change in functional composition along a successional gradient. For some traits, dry and wet forest succession exhibit parallel changes from ‘fast’ to ‘slow’ traits. Other traits show  contrasting trends along the successional gradients which can be explained by the need for drought resistance in early dry forest succession.

 

 

(2) Successional convergence in functional traits and species composition of trees in Costa Rican wet forests

Boukili V1, Chazdon R11University of Connecticut – Ecology and Evolutionary Biology

 

We compared community-weighted mean (CWM) functional traits of trees >10 cm DBH in six secondary forests (10 to 47 years following pasture abandonment) and two old-growth forests. Species composition was monitored annually over 7 (4 plots) or 15 (4 plots) years. Wood specific gravity and six leaf functional traits were measured for species accounting for 80-95% of the basal area of trees in each plot. If environmental filtering drives successional changes in tree species composition, we predicted a shift over time in abundance and basal area from species with traits that enable fast returns on tissue investment early in succession to species with traits that lead to slow economic returns later in succession. Consistent with our predictions, we found a decline in CWM specific leaf area (SLA), and an increase in CWM leaf dry matter content (LDMC), leaf thickness (LT), leaf toughness (LTO), leaf density (LD), and wood specific gravity (WSG), both within and among successional plots over time. We further examined changes in CWM traits for the smaller (DBH below the median) and larger (DBH above the median) proportions of the tree community. The species composition of the smaller trees in the secondary forests showed higher similarity to the old-growth forest species composition than did the larger trees. In general, the smaller trees in each plot had larger leaves with lower SLA, and higher LDMC, LTO, LD, and WSG. These results demonstrate that new tree recruits in secondary forests have traits associated with the slow end of the leaf economics spectrum. We conclude that community assembly of secondary tropical forests is associated with environmental filtering based on leaf and wood traits. Our long-term data clearly show that CWM functional traits of trees in the secondary forest understory are converging with those of old-growth forests in the region.

 

 

(3) Long-term tree population change in secondary rain forests and the effects of stand and population density, functional diversity and phylogenetic relatedness

Finegan B1, Chazdon RL2, Letcher SG31Tropical Agricultural Centre for Research and Higher Education (CATIE), 2University of Connecticut – Department of Ecology and Evolutionary Biology, 3Purchase College, NY, USA – Department of Environmental Studies

 

Community phylogenetic and functional ecology predict a switch from abiotic to biotic filtering during secondary forest succession, while population ecology predicts that density-dependent effects on growth, recruitment and mortality are major drivers of change. These approaches combined are potentially powerful for understanding succession – the occurrence and strength of density dependence should vary with stand net phylogenetic relatedness NRI and functional diversity FD – but the necessary longterm data for hypothesis testing are scarce. On infertile soils in wet lowland northeastern Costa Rica, we analysed population change for 34 dicot tree species using >20 years of measurements, trees >=10 cm dbh in eight 1.0 ha permanent sample plots, overall stand age range 1-49 yr. We predicted that tree size inequality (coefficient of variation cv of tree basal area G) and cv of G increments would increase with age until reduced by declining recruitment R and increasing mortality M. Additionally, that values of these metrics should increase with overall stand density, density of conspecifics and NRI, but decrease with stand FD; recruitment should show the opposite trends. Initial hypothesis testing used generalized linear models. Patterns of cv G and Ginc were varied but often fitted predictions. As predicted, conspecific initial density and initial stand G were negatively related to recruitment rates R and the former, positively with mortality rates M. Also as predicted, functional dispersion was positively related to R and negatively to M but contrary to predictions, NRI was negatively related to M. Results indicate that density dependent  intraspecific competition is a major driver of the initial decades of successional change, but also suggest that the intensity of density dependent effects on growth and demography varies as predicted with FD, but not with NRI.

 

 

(4) Convergence and divergence in alternative successional pathways in Central Amazonia

Williamson GB1,2, Mesquita RCG2,3, Longworth JB1, Bentos TV2, – 1Louisiana State University – Department of Biological Sciences, 2Biological Dynamics of Forest Fragments Project – Instituto Nacional de Pesquisas da Amazônia (INPA) and Smithsonian Tropical Research Institute , 3Instituto Nacional de Pesquisas da Amazônia-INPA – Department of Ecology

 

The impact of different land use histories on plant succession in the Amazon Basin is poorly known. Two different land use types, abandonment after clearcuts of mature forest and conversion to pastures following clearcuts, lead to domination by different pioneer species, Cecropia species and Vismia species, respectively. We measured changes in vegetation structure and composition over 25 years of succession by monitoring woody plants on sites 2-19 years after abandonment annually for 12 years. We found that stem densities exhibited no significant trend during the first 25 years. Basal area in Cecropia transects increased quickly in the first decade, outpacing accumulation in Vismia transects, but the two had converged in basal area after 25 years. Vismia transects were much more variable than Cecropia transects in stem density and basal area, reflecting cohort growth and thinning in the Vismia succession. Species density increased linearly at a much faster rate in Cecropia than Vismia transects, creating a large divergence after 25 years. Dominance by Cecropia decreases more quickly than dominance by Vismia. Whereas Cecropia sites decreased dramatically in Sorenson similarity during the first decade of succession, Vismia sites remained more similar to each other. This study confirms that succession following primary forest clearings in the Amazon exhibits alternative pathways that correspond to prior land use—clearcuts without further disturbance versus clearcuts converted to pastures. The main divergence in the two successions is the extremely slow accumulation of species and persistence of the dominant species over 25 years in abandoned pastures. 

 

 

(5) Beyond chronosequences: spatio-temporal models of successional vegetation change in three Neotropical forests.

Norden N1, Chazdon R2, Bongers F3, Finegan B4, Martínez-Ramos M5, Mesquita R6, Williamson B71Pontificia Universidad Javeriana, 2University of Connecticut, 3Wageningen University, 4CATIE, 5Universidad Autónoma de México, 6Instituto Nacional de Pesquisas da Amazonia, 7Louisiana State University

 

During succession, fluctuations in population dynamics, driven by demographic stochasticity or abiotic variation, may lead to alternative stable states. The existence of different basins of attraction has been associated with neutral processes, but it may also be related to deterministic processes. For instance, land-use before abandonment, or attributes of the landscape matrix may promote shifts in successional trajectories. This issue has been rarely addressed in secondary tropical forests, as most studies are based on chronosequence methods. Space-for-time substitutions permit an analysis of a much larger time frame than would be otherwise feasible. However, chronosequence results cannot be used to directly measure rates of change in vegetation. These limitations may mislead our knowledge about rates of change during secondary succession. Here, we gathered a unique data set combining temporal and chronosequence data from more than 50 stands in Brazil, Costa Rica and Mexico to evaluate uncertainty and variation in successional trajectories over time. We compared these trajectories across stands within regions with those obtained from null models randomizing either the identity or the age of the stand. Within each region, stem density and species richness showed higher variation than expected under a null model, even when controlling for land-use and for age since abandonment. In contrast, basal area exhibited a more homogeneous pattern in all sites except for Brazil. Despite such variation, chronosequence models do capture a general trend of increase in all attributes but stem density. When these trends were compared across stands within regions, individual trajectories showed significantly different rates of change in stem density, basal area and species richness. Together, our results show that  chronosequence models are uncertain. Such uncertainty arose from important variation in successional trajectories, independently of land-use and the landscape matrix. Thus,  neutral processes might be important during community reassembly in secondary forests.

 

 

(6) Successional changes in ecosystem services in wet and dry tropical forests

Balvanera P1, Arreola F1, Luis Daniel Avila1, Barragan B1, Boege K2, Frans Bongers3, Mónica Flores-Hidalgo1, Gavito M1, García-Frapolli E1, Godinez MC1, Hernández O1, Lohbeck M3, Martínez-Ramos M1, Martínez-Yrizar A2, Mora F1, Paz H1, de la Peña E1, Pérez-Salicrup D1, Schondube J1, Stoner K4, Suazo I5, Siddique I6, del Val E11Universidad Nacional Autónoma de México, Mexico – Centro de Investigaciones en Ecosistemas, 2Universidad Nacional Autónoma de México, Mexico – Instituto de Ecología, 3Wageningen University, Netherlands, 4Texas A & M University – Department of Biological and Health Sciences, 5Universidad Michoacana de San Nicolás de Hidalgo – Instituto de Investigaciones sobre los Recursos Naturales, 6Universidad Federal de Santa Catarina

 

Succession after abandonment of pastures and agricultural fields leads to recovery of forest biodiversity and function. The ecosystem services that are provided by these secondary forests, as well as their tradeoffs, change through succession. We hypothesized that these changes differ between wet and dry forest systems. We synthesize information on changes in ecosystem services along succession in contrasting climates. Our experimental plots are found in a tropical dry and a tropical wet forest in Mexico, include different successional ages and have been monitored for the last 10 years. Microclimatic conditions changed dramatically, particularly in the dry forests: the cooler microclimates during succession are valuable for cattle ranching. Soil fertility and overall soil quality improved slightly with succession. We found strong differences in carbon storage and carbon uptake among and within forest types. Wet forests take up much more carbon than dry ones, and within each forest type carbon storage and uptake varies as a result of management history and surrounding matrix. Successional in insect pest regulation were not clear-cut as succession advanced: abundance of lepidopterous larvae increased from pastures to other successional stages, beetle abundance was  highest in pastures, but insectivorous bats, amphibians and reptile richness and  abundance was highest at intermediate stages. Late stages presented higher proportion of the species found in old growth forest. The amount and diversity of non-timber forest  products that can potentially be used increased with succession, but forage decreased, and wood increased (in the rain forest only). We conclude that tradeoffs among services change through succession and differ between wet and dry forest. Future studies will assess how the different stakeholders perceive these tradeoffs and what are the associated opportunity costs and decisions at stake. Our integrated studies highlight the importance of secondary succession for maintenance of ecosystem services in tropical forested landscapes.

 

 

(7) Successional age and light effects on seedling growth in alternative successional sequences in the Central Amazon

Jakovac ACC1,2, Bentos TV1, Mesquita RCG1, Williamson GB3,11Instituto Nacional de Pesquisas da Amazônia – Projeto Dinâmica Biologica de Fragmentos Florestais, 2Wageningen University – Forest Ecology and Forest Management Group, 3Louisiana State University – Dept. of Biological Sciences

 

In central Amazonia, land use intensity leads to different rates of succession along alternative successional pathways characterized by initial dominance of Cecropia spp or Vismia spp. Cecropia stands follow the classical succession with fast rates of species turnover while Vismia stands exhibit much slower rates of change. Aiming to deeply understand the constraints leading to this alternative successional pathways, we studied seedlings performance under Vismia and Cecropia chronosequences. Through a controlled experiment we determined the effects of successional age and photosynthetic active radiation (%PAR) on relative growth rates (RGR) of nine species of shadetolerant tree seedlings in Cecropia and Vismia dominated successions, varying in age from 1 to 20 years. In every stand we applied two paired treatments to evaluate the role of light in limiting seedlings growth: the original canopy intact and the canopy thinned to 65% of the original basal area. Our results showed that in Cecropia stands, RGR and %PAR decreased with increasing successional age, and RGR was limited by light. In contrast, in Vismia stands light levels and seedling RGR were independent of successional age, and RGR did not vary with %PAR and was lower than in Cecropia stands even under similar light levels. Our study demonstrates that seedlings under Cecropia and Vismia successions perform differently in response to successional age and light availability. Based on seedlings responses, Cecropia successions experience environmental changes through time that facilitate species transitions whereas Vismia successions show few changes and maintain a limiting environment to seedling growth. Even 15 years after abandonment, Vismia-dominated stands are still extremely nonfunctional in restoring environmental features essential to further successional development. These results emphasize that secondary forests regenerating after sequential burning events in Central Amazon may require restoration efforts to disrupt the positive feedbacks precluding natural succession. 

 

 

(8) Amazonian secondary forests respond to climatic variation: biomass dynamics

Mesquita RCG1, TV Bentos1, Jakovac AC1,2, Massoca P1, Williamson GB31Biological Dynamics of Forest Fragments Project, Instituto Nacional de Pesquisas da Amazônia (INPA) and Smithsonian Tropical Research Institute, PO Box 478, 69011-970, Manaus, AM, Brazil, 2Forest Ecology and Forest Management Group, Wageningen University and Research Centre, PO Box 47, 6700AA, Wageningen, The Netherlands., 3Dept. of Biological Sciences, Louisiana State University, Baton Rouge, La 70808, USA

 

The study relates biomass dynamics and climatic events in secondary forest chronosequences of Central Amazonia. Trees with DBH ≥ 3 cm were measured annually between 1999 and 2011 in areas with different use histories: 9 transects on areas clearcut and maintained as pasture with prescribed fires until abandonment, and 8 transects on areas clearcut and abandoned without further use. For each transect we determined above ground live biomass (ABGB), annual mortality (MR) and recruitment (RR) rates. Climate data, obtained from a meteorological tower 50 km from the study area, were related to interannual biomass variations (IBV). Four drought events occurred during the study period, and the El Nino of 2009/2010 caused strong local droughts with 7.2 and 20.4 mm rainfall in September and November of 2009, respectively (6 and 14% of the expected average). For clearcut-abandoned areas, IBV was significant only twice: an increase in 1999/2000 (10.4% ±2.9%) following the 1998 El Nino drought and a decrease in 2009/2010 (12.9±7.2%) during an El Nino drought. On these transects, larger dominant pioneer individuals died disproportionately. Higher rainfall deficit during the dry season caused higher MR (F1,9=5.04, R2=0.36, p=0.05) and lower biomass accumulation (F1,9=6.17, R2=0.41, p=0.03) in the subsequent year for clearcut-abandoned areas. In abandoned pastures, IBV were constant throughout the period, (pG-G = 0.394, F10,80 = 0.815), pointing to a possible higher resilience to droughts. RR was constant through time, being higher only in 1999/2000 on clearcut-abandoned areas. Biomass accumulation rates approached zero during drier years, and in 2009/2010 were only moderately negative (-2.1 ±6.6%). Use history affects floristic composition, with some species more resilient than others, with a differential ability to support extreme climate events. Drought can be an important agent, affecting species turnover and influencing secondary forest succession in the Amazon.

 

 

(9) Seed and fruit tradeoffs – the economics of seed packaging in Amazon pioneers

Bentos T1, Mesquita R1, Camargo J2, Williamson B – 1National Institute for Research in the Amazon (INPA) – Dept. of Research in Ecology, 2National Institute for Research in the Amazon (INPA) – Biological Dynamics of Forest Fragments Project

 

The tradeoff between seed mass and seed number per tree has been widely established for different communities and is interpreted as a tradeoff between establishment and dispersal. Here, we analyze tradeoffs among seed and fruit traits of Amazon pioneer tree species as the economics of seed packaging. We hypothesized that seeds per fruit is the pivotal variable such that for a given fruit mass, more seeds per fruit implies lower seed mass which fosters greater seed number per tree. Similarly, for a given seed mass, more seeds per fruit implies greater fruit mass which is associated with lower fruit number per tree. We tested and generally confirmed these relationships for seed and fruit data from 12 species of pioneer trees from the Central Amazon. Notably, seed number was explained only partially by seed mass (R2 = 0.55), but nearly completely by seed mass, fruit mass and fruit number (R2 = 0.94). In addition, we measured seedling densities in light gaps and adults in secondary forests and then used regression analysis to determine if seedling and adult densities could be explained by seed packaging variables, by fruiting duration, or by germination rates. Seedling and adult densities were best explained by fruit number and fruit mass, not by seed number and seed mass. Fruit variables may be more predictive of seedling and adult densities than seed variables if dispersal is more limiting than establishment. The seed mass-seed number tradeoff in plants, generally interpreted as a tradeoff between dispersal and establishment, is only one component of the economics of seed packaging—i.e., how biomass is allocated among both seed and fruit traits across coexisting plant species.