Spatial distribution of canopy gaps in a tropical forest landscape and its influence on the tree community

Lobo, Elena

Permalink

https://hdl.handle.net/2142/46594 Copy

Description

Title

Spatial distribution of canopy gaps in a tropical forest landscape and its influence on the tree community

Author(s)

Lobo, Elena

Issue Date

2014-01-16T17:55:30Z

Director of Research (if dissertation) or Advisor (if thesis)

Dalling, James W.

Doctoral Committee Chair(s)

Dalling, James W.

Committee Member(s)

• Gertner, George Z.
• Fraterrigo, Jennifer M.
• Greenberg, Jonathan A.

Department of Study

School of Integrative Biology

Discipline

Ecology, Evolution and Conservation Biology

Degree Granting Institution

University of Illinois at Urbana-Champaign

Degree Name

Ph.D.

Degree Level

Dissertation

Keyword(s)

• tropical forest
• treefall gap
• power-law
• spatial pattern
• Light detection And ranging (LiDAR)
• Panama

Abstract

Tropical forests are highly dynamic and diverse ecosystems, but our understanding of the processes that shape the local abundance and spatial distribution of the tree species within them is still rather limited. In particular, the role of disturbance in determining compositional variation in forests remains poorly understood (Molino and Sabatier, 2001). This gap in understanding in large part reflects the lack of appropriate tools that allow measurement of disturbance patterns at fine spatial scale (Chambers et al. 2009). In old growth tropical forests, the creation of canopy gaps by one or multiple tree falls is one of the main causes of disturbance and thus potentially one of the main drivers of forest dynamics (Swaine and Whitmore 1988). My thesis focuses on characterizing patterns of gap disturbance across a forest landscape at Barro Colorado Island (BCI), Panama, and its impact on species distribution. First, I addressed the spatial component of canopy gap disturbances, by exploring the patterns of canopy height across the forest landscape. I considered three main factors: the threshold canopy height used to define gaps, the spatial resolution of canopy height measurements and the extent of the study area. In contrast to prior plot-based studies of gap disturbances, I used canopy height estimates generated from high-resolution LiDAR (light detecting and ranging) for the entire landscape of BCI (1500 ha). I found that the two main metrics that define gap disturbances, namely the gap size-frequency distribution, and fraction of area under gaps, were highly variable at the 1-10 ha plot scale leading to potentially unreliable estimates of disturbance regimes. In addition, when I compared gap disturbance parameters generated with LiDAR with simulated field data, I found that simulated field data underestimated the frequency of large gaps, and therefore potentially, the importance of large disturbances that impact landscape-level carbon storage. Furthermore simulated field data generated significant errors in the location of gaps, which may lead to misleading conclusions in the interaction between canopy gaps and tree recruitment. Second, I assessed how variation across the forest landscape of some key factors affects the gap disturbance regime of BCI. Specifically, I used LiDAR to explore how forest age, topography and soil type affect canopy disturbance patterns across the landscape. Using a Bayesian hierarchical model to model the gap size frequency distribution I showed there was a higher frequency of large gaps in areas of old-growth forest and on more gentle slopes than on steeper slopes and in younger forest. Slope and forest age had similar effects on the gap area fraction, however gap area fraction was also affected by soil type and by aspect. I concluded that variation in disturbance patterns across the landscape can be linked to factors that act at the fine scale (such as aspect or slope) and factors that show heterogeneity at coarser scales (such as forest age or soil type). Awareness of the role of different environmental factors that influence treefall gaps can help scale-up the impacts of canopy disturbance on forest communities measured at the plot scale to landscape and regional scales. Third, I assessed how the distribution of canopy heights in combination with slope and soil chemistry correlates with the distribution of tree species across the 50-ha forest dynamics plot on BCI. Association patterns were analyzed after either classifying individuals according to size class, or after classifying species according to mature habit. Additionally, I also explored whether the frequency of associations with topography and soils differs between species groups that show preferences for either high or low canopy height environments. The most striking patterns that emerged were: i) A large fraction of species, were positively associated with low canopy heights, indicating a preference for recently disturbed areas. These included many understory species that were not previously classified as pioneers. ii) A similarly large fraction of species was found to be positively associated with steep slopes; the frequency of associations with steep slope habitats significantly decreased as tree size increased, indicating a possible ontogenetic shift in the strength of species associations with slope. This pattern may be caused by a greater initial recruitment on moister slopes, and a progressive weakening of differential survival as trees grow and are better able to survive drought events. Analysis of the interactions among environmental factors revealed two contrasting groups of habitat associations: i) positive associations with high slopes and high canopy height, and ii) positive associations with low slopes and low canopy heights. Broadly similar patterns were observed when considering individual size or habit classes. In conclusion, canopy gap disturbances play a crucial role in shaping tropical tree communities. To fully account for the effects of canopy gap disturbances as a driving force affecting the composition and spatial distribution of tree communities, we need to consider the different parameters that comprise a gap disturbance regime. These include the parameters used to define canopy gaps (canopy height threshold, spatial resolution of canopy heights and extent of the study area), the factors that cause spatial variation of gap disturbances (slope, forest age and soil type), as well as the interactions among environmental factors (canopy height, slope, soil chemistry) and their effects on the tree community.

Graduation Semester

2013-12

Permalink

http://hdl.handle.net/2142/46594

Copyright and License Information

Copyright 2013 Elena Lobo

Owning Collections