Three-dimensional forest canopies and their spatial relationships to understory vegetation

Forest canopy structure is one of the most critical structural components involved in the maintenance of ecosystem function. Despite growing appreciation for the ecological role canopies play, advances in our knowledge of canopy structures have been slow relative to other areas of forest ecology. A spatially explicit model, parameterized by intensively sampled field observations, was used to construct the crown and stem-defined forest structure. Crown shells were modeled based on crown dimension information from field observations. Canopies were represented by adding unique crowns to stem locations of trees. Canopy patches were delineated at multiple horizontal and vertical scales using the ARC/INFO Geographic Information System (GIS). The modeling approach used in this study has general use in characterizing three-dimensional canopies of many types of forests. One of the contributions of this model is its ability to examine the vertical structure of forest canopies. In the future, this model may serve as an efficient tool for forest planners to view and quantify forest structure, simulate various practices, and examine the effects of silvicultural treatments on forest structure. Nearest neighbor, Ripley's K, and wavelet analyses were used to explore the canopy structure and responses of understory vegetation to overstory canopy structure. Canopy structure influences the spatial distributions of understory vegetation. To quantitatively identify the spatial relationship of canopy structure and understory vegetation at multiple scales, understory vegetation was recorded by placing 2 m x 2 m plots every 5 m along two 400 m transects in an eight ha 500-year-old Douglas-fir stand in the T. T. Munger Experimental Forest, WA. Multivariate analysis and wavelet analysis were used to identify the possible effects of environmental variables on understory vegetation. Pearson correlation results indicate associations of understory species with the three-dimensional canopies. Most dominant herb species were highly associated with canopy openings and were more negatively affected by lower canopy layers than by dominant-codominant layers. Multivariate analysis (detrended correspondence analysis, canonical correspondence analysis, and two-way indicator species analysis) showed heterogeneous responses of different understory variables to canopy openness and disturbance. Wavelet analysis indicated that the relationship between canopy structure and understory vegetation is scale-dependent (i.e., understory variables respond differently to the canopies across scales). The strongest association among relationships between the canopies and understory vegetation, as well as among the variables within each of these two groups, is at scales of 20-40 m in the study area.